Font Size: a A A

Research On 2D Image Enhancement And 3D Entity Reconstruction Based On Machine Vision

Posted on:2020-09-22Degree:MasterType:Thesis
Country:ChinaCandidate:D D LiuFull Text:PDF
GTID:2428330572990879Subject:Power Engineering and Engineering Thermophysics
Abstract/Summary:PDF Full Text Request
Machine vision involves knowledge of camera,lens construction,light source,image processing algorithm,machinery,software and many other aspects.A perfect machine vision system,like an excellent photography,should pay attention to both"early" construction and "late" processing.Preliminary construction refers to the construction of a reasonable detection environment by means of lighting and noise reduction,so as to obtain clear targets and pictures with rich details.Post processing refers to the use of digital image processing technology,two-dimensional or three-dimensional processing of photos,and the extraction of target and target information.Starting from the "early stage" detection environment,this paper carried out theoretical discussion and experimental research on 2d image enhancement and 3d object reconstruction,and completed the design and construction of a set of machine vision system based on the above research.The following is the main research introduction of this paper:(1)machine vision detection environment analysis.Color test,shape test and noise simulation test were carried out to explore the effect of light source and noise on the target object and its imaging effect.(2)spatial and frequency enhancement of two-dimensional images.Taking the gray scale image of human eyes as an example,various enhancement methods of two-dimensional images in spatial domain and frequency domain are discussed,and the enhancement effects of various methods on gray scale images of human eyes are analyzed and compared from the aspects of two-dimensional expression and three-dimensional expression.(3)point cloud acquisition and reconstruction of 3d images.The 3D printing model designed by ourselves is taken as the target.Based on the laser triangulation method,the model is scanned by the linear laser scanning camera.The 3D reconstruction process of the object is discussed in detail.(4)construction and application of machine vision system.To an enterprise production car oil cooler interlayer virtual welding problems,for example,using area of image enhancement,edge detection,identification and other image processing technology,design a set of machine vision system,and the image processing method,hardware system and software system for the construction of the system,system commissioning test,parameter setting,automatic photos,automatic identification,record number,the event log,data backup,and other functions,is now as a human eye detection of auxiliary equipment into the production line.
Keywords/Search Tags:Light source system, Space&Frequency domain enhancement, Feature recognition, Defect detection
PDF Full Text Request
Related items